All Articles  (X)

We are happy to welcome Mostafa Hagog to Modular, who recently joined to lead our high performance numeric kernels, graph compiler, and low level heterogeneous runtime teams! These technology areas are critical low-level components of our AI Engine, and are directly responsible for delivering state of the art performance across many categories of hardware.

We released Mojo local download for Linux systems on September 7th, and in just over a month we’ve seen tens of thousands of downloads and amazing community projects such as llama.🔥and Infermo. We deeply value your feedback, as Mojicians, and we appreciate every single PR, feature request, and bug report.

Jeff, Abdul, and Chris presented a talk about Mojo and how it relates to MLIR and LLVM, at the LLVM Developer Conference 2023.

Mojo SDK was released in September 2023. As someone who relies on the simplicity of Python and also cares about high performance delivered by languages like C, I was excited to try out Mojo. I felt the same joy and thrill I had experienced when I first discovered programming and ran “Hello World” in QBasic and Turbo Pascal.

Mojo allows you to access the entire Python ecosystem, but environments can vary depending on how Python was installed. It's worth taking some time to understand exactly how modules and packages work in Python, as there are a few complications to be aware of. If you've had trouble calling into Python code before, this will help you get started.

How do you guarantee that your software is portable, runs reliably, and scales easily in production environments? The short answer is: Use containers. Container technologies like Docker and Kubernetes are popular tools for building and deploying software applications, but until recently they were considered exotic infrastructure for IT/Ops experts. 

AI systems take an incredible amount of time to build and get right - I know because I have helped scale some of the largest AI systems in the world, which have directly and indirectly impacted billions of people. If I step back and reflect briefly - we were promised mass production self-driving cars 10+ years ago, and yet we still barely have any autonomous vehicles on the road today.

Since our launch of the Mojo programming language on May 2nd, more than 120K+ developers have signed up to use the Mojo Playground and 19K+ developers actively discuss Mojo on Discord and GitHub. Today, we’re excited to announce the next big step in Mojo’s evolution: Mojo is now available for local download – beginning with Linux systems, and adding Mac and Windows in coming releases.

We started this blog post series to describe how to write Mojo🔥 for the Mandelbrot set to achieve over 35,000x speedup over Python. To recap the optimizations so far, in part 1 we ported the code into Mojo to get around a 90x speedup, and then, in part 2 we vectorized and parallelized the code to get a 26,000x. This blog post continues this journey to show another performance technique that takes well beyond our promised 35,000x speedup goal.

In this blog post (part 2 of our 3 part blog post series), we continue our optimization journey and describe how to go from 90x to 26,000x speedup over Python. We will share insights into the techniques we use and discuss why Mojo is well positioned to address them.